Aryanti, Trimulya Oktavia (2025) Pengembangan Timbangan Bayi Terintegrasi IoT Untuk Pemantauan Tumbuh Kembang Dengan Analisis Regresi Logistik. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of BukuTA_noWM.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
BukuTA_noWM.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy |
Abstract
Pemantauan tumbuh kembang bayi selama tahun pertama kehidupan sangat penting untuk mendeteksi dini risiko malnutrisi yang kerap dipengaruhi oleh faktor biologis, lingkungan, dan sosial. Penelitian ini mengembangkan sistem timbangan digital bayi berbasis Internet of Things (IoT) yang dilengkapi dengan analisis regresi logistik sebagai metode prediktif status gizi. Sistem ini memanfaatkan sensor load cell yang terhubung ke mikrokontroler MKR WiFi 1010 melalui modul konversi sinyal ADS1232 24-bit, memungkinkan pencatatan data berat badan secara akurat. Pengujian titik beban menunjukkan posisi tengah sebagai titik paling akurat dengan rata-rata galat sebesar 0,87% dan standar deviasi 0,28% pada beban 500 gram. Untuk pengukuran panjang badan, digunakan sensor rotary encoder yang menghasilkan pulsa digital dan dikalibrasi untuk konversi ke satuan sentimeter. Rata-rata error pengukuran panjang mencapai 0,45%, dan akurasi ditingkatkan melalui koreksi berbasis regresi linier. Sistem ini juga mengintegrasikan data subyektif dari wawancara, seperti demografi, riwayat kesehatan ibu dan bayi, serta kondisi lingkungan. Model regresi logistik biner yang dibangun menunjukkan performa klasifikasi yang baik, dengan sensitivitas 94,44%, nilai prediksi negatif (NPV) 93,33%, dan F1-score sebesar 85,00%. Validasi menggunakan metode Leave-One-Out Cross-Validation (LOOCV) menunjukkan indikasi overfitting awal, yang kemudian diatasi melalui integrasi data sekunder dan teknik Enhanced SMOTE untuk penyeimbangan kelas. Dengan performa sensor yang stabil dan akurasi model prediktif yang tinggi, sistem ini memiliki potensi besar sebagai alat skrining dini risiko malnutrisi pada bayi, serta sebagai pendukung pengambilan keputusan bagi tenaga medis dan keluarga.
=====================================================================================================================================
Monitoring infant growth during the first year of life is essential for the early detection of malnutrition risks, which are often influenced by biological, environmental, and social factors. This study developed a digital baby scale system based on the Internet of Things (IoT), integrated with logistic regression analysis as a predictive method for nutritional status. The system utilizes a load cell sensor connected to an MKR WiFi 1010 microcontroller through a 24-bit ADS1232 signal conversion module, enabling accurate weight measurements. Load point testing showed that the center position provided the most accurate results, with an average error of 0.87% and a standard deviation of 0.28% at a 500-gram load. For body length measurements, a rotary encoder sensor was used to generate digital pulses, which were calibrated and converted into centimeters. The average length measurement error reached 0.45%, and accuracy was improved through linear regression-based correction. The system also integrates subjective data from interviews, such as demographics, maternal and infant health history, and environmental conditions. The developed binary logistic regression model demonstrated good classification performance, with a sensitivity of 94.44%, a negative predictive value (NPV) of 93.33%, and an F1-score of 85.00%. Validation using Leave-One-Out Cross-Validation (LOOCV) initially indicated overfitting, which was addressed through the integration of secondary data and the use of Enhanced SMOTE for class balancing. With stable sensor performance and a highly accurate predictive model, this system holds significant potential as an early screening tool for infant malnutrition risk and as a decision support system for healthcare providers and families.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Pertumbuhan bayi, integrasi IoT, pemantauan antropometri, regresi logistik, timbangan bayi |
| Subjects: | T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.62 Decision support systems T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105 Data Transmission Systems T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments |
| Divisions: | Faculty of Electrical Technology > Biomedical Engineering > 11410-(S1) Undergraduate Thesis |
| Depositing User: | Trimulya Oktavia Aryanti |
| Date Deposited: | 04 Aug 2025 06:39 |
| Last Modified: | 04 Aug 2025 06:39 |
| URI: | http://repository.its.ac.id/id/eprint/126810 |
Actions (login required)
 |
View Item |